Colorectal cancer accounts for approximately 10% of all new cancer cases and 10% of all cancer-related deaths in the United States. In fact, colorectal cancer causes more deaths than prostate cancer in men 60 to 79 years of age and more deaths than breast cancer in women 80 years or older. In 2008, 108,070 new cases of colon cancer and 40,740 new cases of rectal cancer will be diagnosed. The probability of developing colorectal cancer increases from 0.07% in the first four decades of life to 4.3% for females and to 4.8% for males in the seventh decade of life. Although the 5-year relative survival (66%) with this cancer is identical for persons younger than 65 and for persons 65 to 74 years, it declines to 60% for individuals 75 years or older (Figure 98-3 and Figure 98-4).
Six out of seven studies suggest that elderly patients present at the time of diagnosis with the same probabilities of having localized and advanced colon cancer as younger patients. In randomized studies, elderly patients are reported to have the same performance status and incidence of tumor-related symptoms as younger patients, but tend to have more disease-related weight loss. Older patients also appear to have a greater incidence of right-sided (proximal) tumors and higher rates of obstruction and perforation. This in turn results in a greater likelihood of acute presentations, leading to a higher perioperative mortality rate in this older age group.
While some investigators have noted that overall survival is shorter in elderly patients, this difference is not significant if noncancer deaths are excluded. In fact, comorbid illnesses account for a substantial number of deaths among older patients with colorectal cancer. In one study of 29,733 patients 67 years and older with localized colorectal cancer in the Surveillance, Epidemiology and End Results (SEER)/Medicare database, congestive heart failure, chronic obstructive pulmonary disease, and diabetes mellitus accounted for 9.4%, 5.3%, and 3.9% of deaths, respectively.
Risk Factors of Colorectal Cancer
As in younger individuals, most older persons have no clearly defined risk factors for the development of their colorectal cancer. Hereditary syndromes, such as familial adenomatosis polyposis or hereditary nonpolyposis colorectal cancer, have less impact in this older age group, since the majority of cases are diagnosed before the age of 65 years. There is one risk factor specific to men who were curatively treated with radiation for localized prostate cancer. These men have a 70% increased risk of developing cancer in previously irradiated portions of the bowel and could potentially benefit from more frequent colorectal cancer screening.
Chemoprevention of Colorectal Cancer
Various dietary supplements and drugs have been proposed as chemopreventive agents for colorectal neoplasia. Of these, aspirin has been the most widely studied; the regular use of this drug appears to confer a significant risk reduction (relative risk 0.56–0.68).
In at least two prospective cohort studies, the use of postmenopausal estrogens has been shown to reduce the risk of colorectal cancer by approximately 30%. Estrogen supplementation, however, has been associated with an increased risk for coronary heart disease, breast cancer, thromboembolic events, and early mortality in at least one large randomized study of postmenopausal women, thereby limiting its use as a chemopreventive agent in older women.
Most recently, it has been reported that cholesterol-lowering “statins” can significantly reduce the risk of colorectal cancer. As compared to individuals who did not use statins, individuals who used these agents for at least 5 years were able to reduce their relative risk of colorectal cancer by one-half.
Screening of Colorectal Cancer
Approximately one in two persons will have an adenomatous polyp in their colon by the age of 70 years. The risk of sporadic colorectal cancer can be reduced by 50% to 90%, if adenomatous polyps are implicated by fecal occult blood testing or are visualized by barium enema, sigmoidoscopy, or colonoscopy and are then removed before they transform into a malignancy over a 5 to 10 year period. Three randomized trials have demonstrated that routine fecal occult blood testing can diminish colon cancer mortality by 15% to 20% during an 8 to 18 year follow-up period. The mortality benefit for older patients was slightly lower (10%–16%) than for younger patients (19%–23%).
Case–control studies suggest that lower endoscopy and removal of polyps may decrease the incidence and mortality of colorectal carcinoma by 50% to 80%. However, screening extends life to a lesser degree with advancing age. Although colonoscopy appears to be particularly beneficial in older patients who have a higher incidence of proximal neoplasms, which are beyond the view of a flexible sigmoidoscopy, the risk of perforation from colonoscopy is higher than from sigmoidoscopy (OR 1.8). This risk increases with advancing age and number of comorbidities. Therefore, screening decisions must be individualized for every elderly patient.
Since only a small percentage of persons at risk for colorectal neoplasia undergo routine screening in the United States, Medicare now provides reimbursement for surveillance fecal occult blood testing plus sigmoidoscopy or barium enema or colonoscopy for all beneficiaries. This has increased the prevalence of “screening within 1 year” for fecal occult blood testing from 20% in persons younger than 65 years to 26% in persons 65 years and older and the prevalence of “endoscopy within 5 years” from 37% in persons younger than 65 years to 48% in persons 65 years or older.
Most older persons have a stated preference to continue colorectal cancer screening, feeling that their own life expectancy does not factor into this decision. Since the risk of advanced neoplasia and colorectal cancer is highest in the oldest individuals, it seems doubtful that screening should be arbitrarily discontinued after a certain age. In the United States, an individual who reaches the age of 80 years has an average life expectancy of an additional 8.6 years for women and 6.7 years for men. At age 85 years, average survival is 6.7 years for women and 4.7 years for men. This estimate must be adjusted for the number and severity of chronic diseases affecting the individual, as well as his or her functional status. It may be reasonable to discontinue screening when life expectancy is shorter than the time a polyp progresses to a cancer, that is, 5 to 10 years.
Localized Colorectal Cancer
Surgery—Localized Colon Cancer
Cancers of the colon are typically resected with a hemicolectomy. Whether older age is an adverse prognostic factor in resection of localized colon cancer is controversial. Data on 20,862 patients undergoing surgery in 1997 for colon cancer from the Nationwide Inpatient Sample, a claims-based database, suggest that perioperative mortality increases gradually with advancing age until the age of 80 years, after which a substantial increase in mortality is seen (6.9% for patients older than 80 years). In Australia, patients 70 years and older had a worse prognosis for early-stage colon cancer treated with surgery alone. A European study also showed increased 30-day morbidity and mortality rates for patients 80 years and older compared to their younger counterparts.
Older patients in the Nationwide Inpatient Sample from 1997 had lower mortality rates at high-volume hospitals than at low-volume institutions: 3.1% versus 4.5% for patients older than 65 years (p = 0.03). This effect was even more pronounced for patients older than 80 years, where 4.6% of patients died at high-volume centers compared to 7.3% at smaller institutions (p = 0.04). Therefore, it may be preferable to send the very oldest patients with colon cancer to high-volume hospitals for their resections.
A tendency to perform less aggressive surgery in older patients may be evident from a recent analysis of 116 995 adults with localized colorectal cancer in the SEER database (1988–2001). This retrospective study revealed that patients 71 years or older were half as likely to receive adequate lymph node evaluation (examination of at least 12 lymph nodes) as younger patients. Multiple studies have now demonstrated that inadequate lymph node evaluation correlates with inferior survival. These data would suggest that less aggressive surgical intervention predisposes elderly patients to a higher risk of recurrence and cancer-related death.
Increasingly, laparoscopy-assisted colectomies are being performed in the United States and elsewhere. Randomized studies indicate that patients undergoing this procedure have a similar outcome to patients undergoing an “open” colectomy, with slightly less postoperative pain and a one-day shorter hospital stay. Persons 75 years or older tolerate laparoscopic-assisted colectomy equally well as younger individuals. One study compared 51 patients 70 years or older who underwent laparoscopic colectomy to 102 age-matched patients who underwent open colectomy. Older patients had less overall morbidity with laparoscopic colectomy (17.6%) than with open surgery (37.3%), suggesting that laparoscopic surgery may lessen surgical morbidity for frail, elderly individuals (p = 0.01).
Surgery—Localized Rectal Cancer
A radical resection with anastomosis or ostomy is the standard surgical procedure for cancers of the rectum. A Veteran's Administration study of 7243 patients who underwent surgery for their rectal cancer during the period of 1990 to 2000 revealed that 30-day mortality following resection for rectal cancer was 2.1% for patients younger than 65 years compared with 4.9% for patients 65 years or older (p < 0.0001). Five-year survival was 60% versus 48%, respectively (p < 0.0001). French investigators reviewed the records of 92 surgical patients who were 80 years or older and matched these to records of 276 younger patients who underwent resection during the same time interval. Although not statistically significant, older patients had a higher operative mortality than younger patients (8% vs. 4%, respectively). Among elective surgeries, the operative mortality was nearly identical (3%–4%). Although 5-year overall survival was greater for younger patients, 5-year cancer-specific survival was comparable for the two groups.
Adjuvant Therapy—Localized Colon Cancer
Since the publication of a National Institutes of Health (NIH) consensus statement in 1990, 5-fluorouracil (5-FU)-based adjuvant therapy has represented the standard of care for patients in the United States following the complete resection of colon cancer that has spread to regional lymph nodes (stage III disease). A benefit for postoperative chemotherapy has not been prospectively demonstrated in patients with fully resected, muscle-invasive, lymph node-negative (stage II) colon cancer. However, stage II cancers with high-risk features for recurrence (e.g., presentation with obstruction, perforation, or invasion into adjacent organs—stage IIb) are often treated by medical oncologists and are included in many randomized adjuvant trials.
Older patients receive adjuvant chemotherapy less often than their younger counterparts. A retrospective cohort study, utilizing the SEER database, identified 6262 patients 65 years or older with resected stage III colon cancer from 1991 to 1996. Overall, 55% of patients received adjuvant chemotherapy within 3 months of colon cancer resection. The likelihood of receiving treatment declined steeply with increasing age (Table 98-1). Similar results were noted in an Italian study of 1014 patients with resected stage II/III colon cancer.
Table 98-1 Use of Chemotherapy for Colorectal Cancer by Age ||Download (.pdf)
Table 98-1 Use of Chemotherapy for Colorectal Cancer by Age
TOTAL NUMBER OF PATIENTS
PERCENTAGE OF PATIENTS TREATED
HOSPITALIZATIONS FOR CHEMOTHERAPY-RELATED TOXICITY
Stage III colon cancer
Stage II/III rectal cancer
Stage IV colorectal cancer
The reason for this age-related disparity in management is not entirely clear. In one study utilizing the California Cancer Registry, oncologists cited patient refusal, comorbid illness, or advanced age as the most common reasons for not providing chemotherapy to elderly patients with resected colon cancer. Financial considerations and logistical problems may also prevent some elderly from seeking care. Additionally, treatment for older patients is more frequently discontinued, suggesting reluctance by physicians to treat older patients who have experienced some degree of side effects to chemotherapy.
A pooled analysis of 3351 patients in the United States, stratified by decade of life, from seven randomized trials evaluated the benefit of adjuvant chemotherapy in elderly persons with stage II or III colon cancer. The primary conclusion of this pooled analysis was that elderly patients derive the same clinical benefit from postoperative chemotherapy as younger patients. Treatment-related toxicity was somewhat higher for older patients, yet not statistically significant. Another U.S. study, however, could not extend this observation to patients 75 years and older, given limitations in the data.
Adjuvant chemotherapy also appeared to be slightly more toxic for elderly patients in a SEER-derived retrospective cohort study (Table 98-2). Hospitalization for various chemotherapy-related toxicities increased steadily with advancing age. Prospective data from the Royal Marsden Hospital in London confirmed this trend toward higher rates of severe toxicity in patients 70 years or older.
Table 98-2 Severe Toxicity to 5-Fluorouracil-Based Therapy by Age in a Pooled Analysis of 1748 Patients with Advanced Colorectal Cancer Treated by the North Central Cancer Treatment Group ||Download (.pdf)
Table 98-2 Severe Toxicity to 5-Fluorouracil-Based Therapy by Age in a Pooled Analysis of 1748 Patients with Advanced Colorectal Cancer Treated by the North Central Cancer Treatment Group
SEVERE TOXICITY (NCI CTC GRADE ≥3)
In conclusion, it appears that postoperative chemotherapy in stage III colon cancer is as beneficial in elderly patients, as it is in younger individuals. A slight increase in toxicity should not prevent the clinician from treating these older patients, although increased vigilance during chemotherapy is warranted.
Adjuvant Therapy—Localized Rectal Cancer
Chemoradiation before or after surgery reduces the local recurrence rate and increases disease-free survival in patients with deeply invasive (T3–4) or lymph node–positive rectal cancer. Although there are few data to suggest that elderly patients respond any differently to chemoradiotherapy than younger patients, they are less frequently referred to oncologists for this treatment. Analysis of the SEER database identified patients 65 years or older with stage II or III rectal cancer who underwent surgical resection between 1992 and 1996. Increasing age corresponded inversely with the percentage of patients who received adjuvant therapy (Table 98-1). Overall, chemoradiation therapy was associated with a 17% reduced risk of death among all cases, with a 29% reduced risk of death in stage III patients, but no statistically significant improvement in stage II cases, similar to the results reported in younger patients.
Postoperative Surveillance—Localized Colorectal Cancer
There is general agreement that all patients with resected colorectal cancer should undergo regular surveillance screening with colonoscopy, CEA testing, and possibly an annual CT scan. Analysis of 52,283 Medicare beneficiaries treated for locoregional colorectal cancer between 1986 and 1996 suggests that younger patients are more likely to undergo periodic surveillance endoscopies, and that the median time to first follow-up endoscopy is significantly shorter (p < 0.0001) for patients at younger ages.
Advanced Colorectal Cancer
Surgery—Advanced Colorectal Cancer
Removal of the primary cancer is indicated in patients who have an (impending) obstruction, uncontrollable bleeding, or oligometastatic disease that may potentially be cured with aggressive therapy. In spite of these restrictive criteria, a recent pattern of care study, utilizing the SEER database, revealed that the majority of patients 65 years or older with stage IV colorectal cancer undergo resection of their primary tumor. Overall, 72% of patients underwent this primary-cancer-directed surgery. The percentage of patients undergoing surgery declined gradually from 76% in patients 65 to 69 years old, to 70% in patients 80 to 84 years old, and then dropped to 62% in patients 85 years or older.
In this pattern of care study, the 30-day mortality was 10% for all patients. However, higher perioperative mortality rates for elderly patients with advanced colorectal cancer have been reported elsewhere; in patients 70 to 79 years of age, the mortality may be as high as 21% and in octogenarians it may reach 38%. Overall, improvements in surgical technique and postoperative care have led to a decrease in operative mortality particularly in older patients, so that current operative survival figures for elderly patients now approach those reported for younger patients a decade ago.
Elderly patients with oligometastatic disease or isolated recurrences of their colorectal cancer may be candidates for a “curative-intent” resection. In a series from the Memorial Sloan-Kettering Cancer Center, 128 patients 70 years or older underwent liver resection for metastatic colorectal cancer between 1985 and 1994. While these patients experienced a 4% perioperative mortality rate and a 42% complication rate, their median survival was 40 months and 5-year survival rate was 35%. These older patients had a similar outcome to 449 patients younger than 70 years who underwent comparable liver resections during the same time period. In a retrospective analysis of patients 65 years and older in the SEER database, only 6.1% of the 13,599 patients identified with colorectal metastases limited to the liver underwent hepatic resection. The 30-day mortality rate was 4.3%. Five-year survival for resected patients was superior to those who did not undergo resection (32.8% vs. 10.5%; p < 0.0001). Additionally, investigators have reported that appropriately selected elderly patients can tolerate resection of colorectal lung metastases with similar outcome as younger patients, with acceptable morbidity and mortality.
Chemotherapy—Advanced Colorectal Cancer
Until the early 1990s, the only active treatment for advanced colorectal cancer was 5-FU, a thymidylate synthase inhibitor. Multiple studies have now demonstrated that this agent, given as a bolus with leucovorin or as an infusion with or without leucovorin, is effective and well tolerated in the elderly population, with similar response rates and overall survival compared to younger cohorts.
Folprecht and colleagues carried out a pooled analysis of 22 randomized European trials in which patients received palliative 5-FU-based therapy. In this retrospective analysis, overall survival (10.8 months for younger patients vs. 11.3 months for older patients, p = 0.31) and response rate (20.8% for younger patients vs. 17.6% for older patients, p = 0.14) were nearly identical in younger and older patients. Infusional 5-FU proved to be superior to bolus 5-FU in all age groups. Similar results were obtained in a meta-analysis of four large trials of 5-FU-based therapy to examine the efficacy and toxicity of 5-FU in patients 70 years and older (n = 303) compared to younger individuals (n = 1181). While severe neutropenia (40.4% vs. 33.6%) and stomatitis (10% vs. 4.6%) were significantly increased in older patients, overall response rate (39.5% vs. 33.1%) and overall survival (15.9 vs. 15.4 months) were similar to that of younger patients.
The North Central Cancer Treatment Group (NCCTG) conducted another pooled analysis of 1748 patients with advanced colorectal cancer treated with 5-FU-based therapy. No significant differences in response rate (p = 0.90) or overall survival (p = 0.42) were observed between patients younger than 56 years and older than 70 years. Patients 65 years or older had more overall severe toxicity than their younger counterparts (53% vs. 46%, p = 0.01). Statistically significant differences in severe diarrhea and stomatitis were reported. Other studies have also noted higher rates of 5-FU-related palmar–plantar erythrodysthesia in older patients.
The oral 5-FU prodrug, capecitabine, has similar efficacy to monthly intravenous 5-FU and leucovorin (Mayo Clinic Regimen) in the metastatic and adjuvant setting. In patients 70 years or older with metastatic colorectal cancer, the overall response rate with capecitabine was 24% and the overall survival time was 11 months. In patients 70 years and older with resected colon cancer, capecitabine offered the same disease-free survival advantage as standard 5-FU and leucovorin. In this adjuvant trial, severe capecitabine-related toxicity was noted in only 12% of patients. Overall, the results with capecitabine in older patients appear similar to those seen in younger individuals.
Irinotecan, a topoisomerase I inhibitor, is another active drug in metastatic colorectal cancer. A multi-institutional phase II study found similar rates of severe toxicity in patients 65 years or older and those younger than 65 years receiving a weekly schedule of this agent. In a trial comparing the weekly and the tri-weekly dosing regimens of this drug as second-line therapy for metastatic disease, more than one-third of 291 patients were at least 70 years of age. Age greater than 70 years did not affect survival or time to progression, but was associated with an increased risk of severe neutropenia and diarrhea compared to patients younger than 70 years of age.
Various trials have evaluated irinotecan-based chemotherapies in elderly patients with metastatic colorectal cancer. These have included combinations of irinotecan, bolus 5-FU, and leucovorin; irinotecan, infusional 5-FU, and leucovorin; irinotecan and capecitabine; and irinotecan and oxaliplatin. These trials uniformly revealed that response to chemotherapy, median survival times, and degree of toxicity in these elderly patients appear to be similar to those observed in younger patients.
Oxaliplatin is a third-generation platinum analogue, which induces platinum-DNA adducts, inhibiting the replication of DNA. Although it has limited activity as a single agent in colorectal cancer, oxaliplatin has notable synergy with 5-FU and leucovorin (FOLFOX). In metastatic colorectal cancer, the toxicity following treatment with FOLFOX was evaluated for patients 70 years and older in a pooled analysis of four randomized clinical trials. Although severe hematologic toxicity was increased in older patients, there was no difference in other severe toxicities or 60-day mortality.
For patients with resected colon cancer, a randomized adjuvant trial, in which 35% of patients were 65 years or older, has suggested that disease-free survival can be enhanced if oxaliplatin is added to 5-FU and leucovorin. Subgroup analysis suggested a similar benefit for elderly patients as for the group as a whole. For those older patients with resectable liver metastases, preoperative FOLFOX therapy may be superior to 5-FU therapy. In this study, 2-year overall survival was prolonged in the “FOLFOX group” compared to the “5-FU group” or to the “no chemotherapy group” (84% vs. 60% vs. 23%).
The combination of capecitabine and oxaliplatin (CAPOX) has been evaluated in 76 patients 70 years and older with metastatic colorectal cancer. Overall response was 41%, median progression-free survival was 8.5 months and median overall survival was 14.4 months with little severe toxicity (3% peripheral neuropathy and 13% palmar–plantar erythrodysthesia). A recent randomized study has suggested that the FOLFOX and CAPOX regimens have similar efficacy and rates of toxicity.
The humanized monoclonal antibody, bevacizumab, binds to the vascular epidermal growth factor, reducing the vascularity of tumors, leading to hypoxia and necrosis. It also appears to lower interstitial fluid pressure in cancer nodules, promoting diffusion of other chemotherapeutic agents into these tumors. Three pivotal studies in patients with advanced colorectal cancer have established that this agent consistently enhances the efficacy of 5-FU-based therapy with either irinotecan or oxaliplatin for patients of all age groups. A fourth study compared bolus 5-FU and leucovorin alone or in combination with bevacizumab in patients 65 years and older or in patients of poor performance status. This study demonstrated that bevacizumab could significantly improve survival by 4 months in “suboptimal” patients without significantly increasing toxicity.
Bevacizumab may increase the risk of arterial thrombosis, i.e., myocardial infarction, stroke, or peripheral arterial thrombotic event. In a multivariate analysis, persons 65 years or older increased their risk of arterial thrombosis from 2.3% with chemotherapy alone to 7.1% with chemotherapy plus bevacizumab. This risk increased further to 17.9%, if the patient had a history of prior arterial thrombosis. However, progression-free and overall survival was better for those receiving the bevacizumab combination than for those receiving chemotherapy alone. Therefore, bevacizumab can still provide a benefit in carefully selected elderly patients with a history of atherosclerosis-related disease.
Cetuximab and panitumumab are monoclonal antibodies to the epidermal growth factor receptor. Blockade of this receptor in colorectal cancer cells induces apoptosis and cell death. In individuals with extensively treated metastatic colorectal cancer, these agents can induce significant tumor regressions in 8% to 11% of patients; cetuximab in combination with irinotecan has a response rate of 23%. The most common severe toxicities of these agents include dyspnea, asthenia, and an acneiform rash. Cetuximab may also cause a hypersensitivity reaction. In one phase II study of 114 patients 65 years and older with metastatic colorectal cancer refractory to irinotecan, oxaliplatin, and 5-FU, cetuximab achieved a response rate of 9.6% compared to a 12.5% response rate in 232 patients younger than 65 years (p = nonsignificant). In a randomized trial for patients with chemotherapy-refractory metastatic colorectal cancer, panitumumab resulted in partial response rate of 8% and a 46% reduction in risk of tumor progression compared to best supportive care. Similar to cetuximab, rash is the predominant adverse effect of panitumumab therapy. A severe rash early in the treatment course is predictive of enhanced response and survival with either agent.